Combining Spectroscopic and Potentiometric Approaches to Characterize Competitive Binding to Humic Substances

HAL Id: cea-00311465

https://hal-cea.archives-ouvertes.fr/cea-00311465

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Distributed under a Creative Commons Attribution - NonCommercial - NoDerivatives| 4.0 International License

Combining Spectroscopic and Potentiometric

Approaches to Characterize Competitive Binding to

Humic Substances

Laura Marang, Pascal E. Reiller, Sascha Eidner, Michael Kumke, Marc

Benedetti

To cite this version:

Laura Marang, Pascal E. Reiller, Sascha Eidner, Michael Kumke, Marc Benedetti. Combining Spec-troscopic and Potentiometric Approaches to Characterize Competitive Binding to Humic Substances. Environmental Science and Technology, American Chemical Society, 2008, 42 (14), pp.5094-5098. �10.1021/es702858p�. �cea-00311465�

1

Combining Spectroscopic and Potentiometric

Approach to Characterize Competitive Binding to

Humic Substances.

LAURA MARANG,a,b,† PASCAL E. REILLER,a SASCHA EIDNER,c MICHAEL U. KUMKE,c MARC F. BENEDETTI b,*

CEA Saclay, Nuclear Energy Division, DPC/SECR, Laboratoire de Spéciation des Radionucléides et des Molécules, Bâtiment 391 – P.C. 33, BP 11, F-91191 Gif sur Yvette, France, Laboratoire de Géochimie des Eaux, Université Paris Diderot, IPGP and UMR CNRS 71574, Case Postale 7052, 75251 Paris Cedex 05, France, Institute of Chemistry, University of Potsdam,

Karl-Liebknecht-Straße 24-25, 14476 Potsdam-Golm, Germany

*

Corresponding author fax: +33 1 44 27 60 38; e-mail, benedetti@ipgp.jussieu.fr

a

Laboratoire de Spéciation des Radionucléides et des Molécules

b

University of Potsdam

c

Université Paris Diderot

†

Present address: Groupe Evaluation des Risques Environnementaux et Sanitaires, EDF R&D, Département LNHE, 6 Quai Watier, F-78401 Chatou, France.

2

ABSTRACT

In an area that contains high concentrations of natural organic matter, it is expected that it plays an important role on the behavior of rare earth elements (REE), like europium, and of trivalent actinides. Competitive interactions with H+, inorganic species, major cations, e.g., Ca(II) or Mg(II), could influence these metals transport and bioavailability. Competitive experiments between cations, which can bind differently to humic substances and Eu3+, will bring an improved understanding of the competitive mechanisms. The aim of this study is to acquire data for Eu(III)/Cu(II) and Eu(III)/Ca(II) competitive binding to a sedimentary originated humic acid (Gorleben, Germany). The NICA-Donnan parameters for Ca2+, Cu2+, and Eu3+ obtained from competitive binding experiments using Ca2+ or Cu2+ ion selective electrodes were used to model time-resolved laser fluorescence spectroscopy (TRLFS) measurements. Eu3+ and Cu2+ are in direct competition for the same type of sites, whereas Ca2+ has an indirect influence through electrostatic binding.

KEYWORDS humic acid, competition, rare earth elements, lanthanides, TRLFS

BRIEFS: Competitive binding between a rare earth analogue and divalent cations onto humic substances reveals different binding sites for the rare earth elements..

3

Introduction

Natural organic matter (NOM) plays a major role on the geochemical cycle and transport of major and trace elements in natural and contaminated environment. The knowledge of radionuclides transport in the geosphere is a key issue for the assessment of the safety of nuclear facilities. Humic acids (HA) and fulvic acids (FA) provide excellent analogues for the reactive components of NOM in soils and rivers and can be used to model metal ion binding to NOM. Several f-block elements, which have a water stable +III redox state, are long-lived radionuclides (RN) typically the actinides (An) like Pu, Am and Cm, and among the lanthanide (Ln) the 151Sm isotope. Moreover, analogies between An3+ and Ln3+ ions are sometimes used to implement database. Complexation of Eu(III) with HA or FA can be considered as an analog to actinide complexation to NOM as well as a relevant test element for a better understanding of rare earth elements (REE) behavior in complex geochemical environment, i.e., soil solution, soils, and aquifers.

Competitive interactions with H+, inorganic species, and major cations (Ca2+, Mg2+…) could influence REE transport and bioavailability. Hence, it is important to obtain reliable Eu3+ binding data for organic and inorganic ligands and to calibrate models in order to predict Eu3+ speciation in natural multicomponent systems (i.e., soils and rivers). Data sets for Eu3+ binding to HA and FA are available for different salinities and pH values (1-4). However, these data sets are frequently not interpreted within consistent modeling framework approaches. Moreover, studies integrating competitive binding effects are scarce. These competition mechanisms between classes of cations (5-7), whereas others are not (8-11), can be interpreted in the framework of NICA-Donnan model (12-14). Competitive experiments between Eu3+ and cations that can bind differently to humic substances could complement the available information. (12)

Since the work from Milne et al. (13, 14), NICA-Donnan generic data for HA and FA proton and metal binding are available. It has been shown that these data can be used, in a first approximation to describe the complexation comportment of various humic samples (15, 16), even if specific data are often still needed in order to increase the confidence in both the model prediction and of the

4 generic data (17, 18). The aim of this work is to obtain laboratory data that can be use to implement models which can help in understanding environmental situation. This is done under experimental conditions limiting the formation of other Eu(III) complexes (e.g., Eu(OH)n3-n and Eu(CO3)m3-2m) in

order to facilitate the implementation of the binding parameters in the models. We present data for Eu3+/Cu2+ and Eu3+/Ca2+ competitive binding to the Gorleben humic acid (Gohy-573 HA). Ca2+ and Cu2+ were chosen because they are known to bind differently to HS (14), i.e., nonspecific versus specific binding for Ca2+ and Cu2+, respectively. The competitive binding experiments using Ca2+ or Cu2+ ion selective electrodes (ISE) are used to derive the Ca2+, Cu2+, and Eu3+ parameters for the Gohy-573 HA, extracted from the German repository test site (19), within the NICA-Donnan approach. Then, these parameters are used to model Time Resolved Laser Fluorescence Spectroscopy (TRLFS) measurements. Based on these results the competitive effect of Ca2+ and Cu2+ on Eu3+ speciation is predicted.

Materials and Methods

Humic Acid Sample. The Gohy-573 HA was extracted from one of the deep groundwaters in the

Gorleben area and was provided by Manfred Wolf (Institut für Grundwasserökologie, GSF – Forschungszentrum für Umwelt und Gesundheit, Munich, Germany). Its isolation, purification and characterisation are described in detail elsewhere (19, 20).

Reagents. The reagents used were Cu(NO3)2 (Certiprep, Spex), Eu(NO3)2 (europium ICP

standard solution, Aldrich), KNO3 (Panreac), KCl (Suprapur, Merck) for ISE experiments. In the

TRLFS experiments the reagents used were Cu(NO3)2, Ca(NO3)2 (Certiprep, Spex), EuCl3 (Aldrich)

and deionized water (Milli Q, Millipore).

Ion Selective Electrode Measurements. The pH was measured using a pH electrode (Metrohm,

6.0133.100) and an Ag/AgCl glass reference electrode (Metrohm, 6.0733.100). After addition of acid or base, the rate of drift for both electrodes was measured after 1 min and readings were accepted when the drift was less than 0.25 mV min-1. For each data point, the maximum drift monitoring time was 20 min. The pH electrode was calibrated with standard buffers (SCHOTT) of

5 different pH (4.01, 6.87, and 9.18, respectively). During the fixed pH experiment, the pH was controlled with CO2 free KOH (0.01 M) and HNO3 (0.01 M).

Cu2+ and Ca2+ Binding to Gohy-573 HA in the Absence and Presence of Eu3+ at pH 5.5.

Performing a speciation calculation without HA, only Eu3+, Ca 2+

, and Cu2+ ions are the dominant species in solution under our conditions (21, 22). The free Cu2+ and Ca2+ concentrations in solution were measured potentiometrically using a Cu2+ ISE (Metrohm, 6.0508.140), and a Ca2+ ISE (Metrohm, 6.0508.110), respectively. Routine calibrations were performed at pH 5.5 in the absence of HA, in the ranges 0.5 μM ≤ [Cu(II)]total ≤ 1 mM in 1 mM KNO3 background electrolyte, and 0.5

μM ≤ [Ca(II)]total ≤ 1 mM in 1 mM KCl background electrolyte.

First, the Cu2+ and Ca2+ parameters were obtained with direct titrations of Gohy-573. Solutions of Gohy-573 HA (60 mg/L for Cu2+, and 200 mg/L for Ca2+) was titrated to pH 5.5, and then pH was kept constant for 30 min to stabilize the Gohy-573 HA before titration with (i) Cu2+ in the concentration range of 10–75 μM, and (ii) Ca2+ _{in the concentration range of 9–200 μM.}

Second, the Eu3+ parameters were obtained through competitive experiments with Cu2+ and Ca2+. Solutions of Gohy-573 HA (60 mg/L) containing 20 μM of Cu2+

, and Gohy-573 HA (200 mg/L) containing 30 μM of Ca2+

, was titrated to pH 5.5 and then pH was kept constant for 30 min before titration with Eu3+ in the concentration range of 3–60 μM in the case of Cu2+, and 1–50 μM in the case of Ca2+.

At each step, the concentration of free M2+ was measured after the pH has been stabilized at pH 5.5 for 10 min and the drift of the electrode was less than 0.1 mV/min. The concentration of free Cu2+ in the solution was then corrected from the dilution factor (less than 10 %), and the amount of Cu2+ bound to Gohy-573 HA was calculated as the difference M(II)total – M2+free.

Time-Resolved Laser Fluorescence measurement. The free europium concentration in solution

was measured using TRLFS. All luminescence measurements were performed at ambient temperature. In the TRLFS experiments a wavelength tuneable Nd:YAG/OPO system (Spectra Physics/GWU) operating at 20 Hz was used as excitation light source. The spectra were recorded

6 with an intensified CCD camera (Andor Technology) coupled to a spectrograph (MS257, Oriel Instruments) as detector. For the time-resolved detection the luminescence signal is collected over a certain time interval (gate width) a certain time (gate delay) after excitation by the laser flash. In the experiments the gate delay was set to 10 µs and the gate width to 100 µs. To increase the signal-to-noise ratio, every spectrum for a certain time step results from an accumulation of 10000 single spectra. Europium (III) is often excited at λex = 394 nm which corresponds to its highest resonant

absorption transition. We decided to excite the samples at λex = 356 nm and λex = 361.7 nm. At λex

= 361.7 nm the luminescence spectra contains significant contributions from both free and bound Eu3+ due to efficient indirect excitation via sensitization by the ligand (antenna effect). The luminescence observed corresponds to the 5D0→7F2 hypersensitive transition and the 5D0→7F1

transition of Eu3+. At λex = 356 nm, the signal of free Eu3+ in water is negligible under our

experimental conditions, and the spectra correspond to Eu3+ bound to Gohy-573 HA. The luminescence spectra were analyzed according to a factor analysis assuming two species. The input parameter were (i) the spectra of a given Eu3+ concentration totally bound to the Gohy-573 HA obtained at λex = 356 nm and (ii) the spectra of free Eu

3+

in water for the same total Eu3+ concentration. Before the fit, all spectra obtained in the presence of Gohy-573 HA in solution were corrected from the absorption of Gohy-573 HA. The percentage of free Eu3+ solution could be obtained from this fit with an error of 10%.

Eu3+ Binding to Gohy-573 HA in the Presence of Cu2+ and Ca2+ at pH 5.5. Competition

experiments between Eu3+ and Cu2+ were made at a Gohy-573 HA concentration of 20 mg/L and a total Eu3+ concentration of 7 μM. The background electrolyte concentration was equal to 1 mM of KNO3 and the pH was set to 5.5 ± 0.1. The total Cu2+ concentration ranged from 1 μM to 0.3 mM.

Competition between Eu3+ and Ca2+ were made with a Gohy-573 HA concentration of 15 mg/L and a Eu3+ concentration of 5 μM; pH was set at 5.5 ± 0.1. The total Ca2+ concentration ranged from 1 μM to 0.1 M. For total Ca2+ concentrations lower than 0.3 mM the background electrolyte was

7 fixed with 1 mM KNO3. For total Ca2+ concentrations ranging from 0.3 to 20 mM, it was fixed with

0.1 M KNO3. For total Ca 2+

concentrations higher than 20 mM the ionic strength was fixed by the Ca2+ salt added in the system and calculated for each data points.

Modeling. Chemical calculations were made using ECOSAT (23), which includes speciation with

inorganic ligands and the NICA-Donnan model, abundantly described elsewhere (12), which describes complexation of metals and protons with humic substances. Deviation from nonideality for inorganic species is accounted with the Davies equation.

Results and Discussion

ISE Data: Ca2+ and Cu2+ Binding to Gohy-573 HA in the Presence and Absence of Eu3+. The

Cu2+ and Ca2+ potentiometric titrations are shown in Figure 1. The amounts of Cu2+ and Ca2+ bound are, in both cases, in the range of previously published data for humic acids of different origin (12, 14). However, the steeper slope of the Ca2+ binding isotherm, compared to the one obtained in a previous study (24), indicates that the Gohy-573 HA is more chemically heterogeneous than the purified peat humic acid, because the degree of heterogeneity controls the slope of the binding isotherm (12).

The effect of Eu3+ is more significant on Cu2+ than on Ca2+ binding (Figure 2). Indeed, for Eu3+ added to the Cu2+ solution with final Eu3+ concentrations ranging from 0 to 60 μM, the percentage of free Cu2+ measured in solution increases from 0 to 25% of the total Cu2+ amount corresponding to an increase of the free Cu2+ in solution by a factor of 10. Whereas in the experiments with Ca2+, for an Eu3+ concentration ranging from 0 to 60 μM, the percentage of free Ca2+ measured in solution varies only by a few percent: the [Ca2+]free in solution is increased by a factor of 1.5. Experiments

with higher Eu3+ concentrations were not performed because they would be irrelevant for most natural or repository conditions.

TRLFS data: Eu3+ Binding to Gohy-573 HA in the Presence of Ca2+ and Cu2+. The effect of

8 Ca2+ (Figure 3). The addition of Cu2+ concentrations ranging from 0 to 0.4 mM increases [Eu3+]free

measured in solution by a factor of 6. Whereas, in order to have the same effect with Ca2+ on the Eu3+ binding, it is necessary to raise the Ca2+ concentration up to 0.1 M. The competitive Ca2+ experiments done with an ionic strength fixed to 1 mM, corresponding to low calcium concentrations, have no effect on Eu3+ binding to Gohy-573 HA. From these competitive experiments, we can conclude that competition between Cu2+ and Eu3+ is more important than for Ca2+ even at high concentrations. These differences are accounted for if Cu2+ and Eu3+ compete for the same binding sites within a given range of total Cu2+ and Eu3+ concentrations. The Ca2+ data show that nonspecific binding is negligible in the case of Eu3+ even at low ionic strength.

Modeling: Specific NICA-Donnan Gohy-573 HA Parameters for Eu3+, Cu2+ and Ca2+

Derived from ISE Experiments. Using the generic parameter values for Eu3+, Cu2+, and Ca2+ parameters (14), a reasonable description of the competitive effect of Eu3+ on Ca2+ and Cu2+ binding to Gohy-573 HA could be achieved (data not shown). However, to have an accurate description of the actual competitive processes, we have determined (Table 1) the specific Gohy-573 HA NICA-Donnan parameters for Ca2+, Cu2+, and Eu3+, and proton parameters were taken form ref (17). The best-fitted ISE data result in an increase in log K_i,Cu2+

~

for both types of sites and to decrease log

+ 3 Eu , 1

K~ for the carboxylic type of sites compared to the generic values in (14). The Ca2+ ISE

experiments could be described without changing the generic parameter values, since binding occurred mostly in the Donnan phase (non-specific binding) for this low ionic strength.

The specific parameters can be used to simulate the speciation of each cation in the Gohy-573 HA (Figure 5) for the experimental conditions applied in the TRLFS measurements. The calculations were done for two ionic strengths in case of Eu3+ (1 mM and 0.1 M), for one ionic strength in case of Cu2+ (1 mM), and to 0.1 M for Ca2+. For Cu2+ and Eu3+, binding occurs mostly at the carboxylic type of sites. Thomason et al. have investigated the binding of Eu3+ to aquatic humic substances using lanthanide ion probe spectroscopy (25). The results, which refer to pH 3.5, showed that at the lowest europium loadings approximately four humic ligands atoms where involved in metal binding.

9 At higher loadings, fewer ligand atoms were involved. The authors consider a series of mono-, bi-, tri-, and tetra-dendate binding sites, probably due to carboxylate groups. In the NICA-Donnan model, the ratio between the non ideality parameter for metal and proton (nM/nH ) per type of sites is

equivalent to the experimental value of the proton-metal exchange ratio (12), i.e., the number of protons expelled for each metal bound to HA. The values of n1,Eu/n1,H = 0.75 and n2,Eu/n2,H = 0.55

suggest that Eu3+ binds most likely in bidentate form. This conclusion is supported by the study of Shin et al. (26) where two contributions were fitted after excitation in the 5D0→7F0 band suggesting

that humic and fulvic acids have two different chemical environments for the binding of Eu3+. These two environments were proposed as phthalate like sites but could also be salicylate. Cabaniss (27) measured synchronous fluorescence spectra of humic extracts and noted that the quenching of the intrinsic fluorescence by Co2+, Cu2+, and Pb2+ were similar at pH 5 and 7.5 suggesting that these metal ions may bind to the same type of sites.

The fitting of the Ca2+ data (Figure 1) is poor. It is important to give a reasonable description of the binding due to the nonspecific interaction and because the Donnan model was proved to be relevant to account for ionic strength effect for HA (28). The most likely reason is a poor knowledge of the actual ionic strength as at very low background electrolyte concentration a small underestimation strongly influences the model outputs. The pH was fixed by adding small amount of KOH and HNO3; this increases the ionic strength of the solution (calculation not shown here).

The ionic strength being higher than expected, i.e., 1.5 mM, a more accurate description of the Ca2+ binding is obtained (Figure 1).

Predicting Ca2+/Eu3+ and Cu2+/Eu3+ TRLFS Data. One of the purposes of modeling is the prediction of metal ion speciation in solution in the presence of competing cations. The direct measurements of all free metal ion concentrations can be difficult and time-consuming. Consequently, it is important to check the ability of the NICA-Donnan model to predict the metal ion speciation in a competitive environment. The free Eu3+ concentrations were determined in TRLFS experiments as a function of the total Ca2+ or Cu2+ concentration and were used to validate

10 the NICA-Donnan parameters (Figure 3, Table 1). Figure 4 presents the free Eu3+ in solution measured by TRLFS vs the free Eu3+ calculated by the NICA-Donnan model. Both are in good agreement as the data points cluster along the one to one line meaning that the increasing competitive effects with the increase of the total Cu2+ or Ca2+ concentration are accurately predicted. The model however overestimates the free Eu3+ in solution for total Cu2+ concentration higher than 0.08 mM (Figure 3 and Figure 4). This could be induced by the lack of data points at higher pH values that are needed to adjust log K2

~

corresponding to phenolic type of sites site for

both cations. Here, generic logK_2,Eu3+

~

for phenolic type of sites were used (14) without any further

adjustment, whereas logK_2,Cu2+

~

was adjusted. From the estimation (29), for high Cu2+ concentration

the proportion of phenolic type of sites should be as important as for the carboxylic ones under our conditions. An underestimation of the logK_2,Cu2+

~

could result, because of a stronger Cu2+

competitive binding, in an increased calculated concentration of free Eu3+ in solution. Further works in order to obtain data obtained at higher pH under different metal to ligand ratio and Cu2+ concentrations are needed to test this hypothesis.

For Ca2+ competitive experiments, the model describes very well the experimental data even when the ionic strength is controlled by the Ca2+ salt added in the system (0.1 M < I < 0.3 M) (Figure 3 and Figure 4), mainly because the interaction of Ca2+ occurs through electrostatic effects. The agreement between experimental and calculated speciation therefore validates the NICA-Donnan parameters used for the simulation of competitive binding.

Implications on the Mobility of REE and Actinides (III). Eu3+, which could be considered as analog for other Ln3+ and An3+ ions, is strongly bound to Gohy-573 HA. Its mobility maybe determined by the colloidal transport, in particularly by NOM as evidenced in ref (30). Ln3+/An3+ -NOM colloids should be less affected in terms of Ln3+/An3+ release in case of a transport through a Ca2+-rich aquifer (calcareous soil or solution, saline solution) since no desorption is expected because of the lack of competition between (Ln3+/An)3+ and Ca2+. Nevertheless, variation of colloid

11 stability and particularly of humic colloid size (31, 32) with Ca2+ and Mg2+ concentration would then be a limiting factor of humic colloid-borne migration of europium. Nevertheless, Ln3+/An3+ fate can be affected in presence of trace elements with high affinity for carboxylic type of sites, e.g., Pb2+ and Zn2+, and also by Cu2+ under low pH conditions (14). In such cases significant amounts of Ln3+/An3+ can be released in solution as free metal and be more available for biouptake. An even stronger effect can be expected in case Al3+ or Fe3+ are present in solution since the trivalent cations would compete for the same sites (11, 16). Further work is needed in order to refine the analysis at higher pHs where the phenolic type of sites should be more influent and where side reactions leading to the formation of either Eu(OH)n3-n or Eu(CO3)n3-2n will take place.

Acknowledgments

This work was financed through “FUNMIG project” (EC: FUNMIG- NUWASTE-2004-3.2.1.1-1), and the MRTRA project of the Risk Control Domain of CEA (CEA/DEN/DDIN). We thank Badia Amekraz, Francis Claret, Christophe Moulin (CEA), and Gunnar Buckau (INE) for their strong support and helpful discussions.

References

(1) Norden, M.; Ephraim, J. H.; Allard, B. The binding of strontium and europium by an aquatic fulvic acid - ion exchange distribution and ultrafiltration studies. Talanta 1993, 40, 1425-1432. (2) Torres, R. A.; Choppin, G. R. Europium(III) and americium(III) stability constants with humic acid. Radiochim. Acta 1984, 35, 143-148.

(3) Lead, J. R.; Hamilton-Taylor, J.; Peters, A.; Reiner, S.; Tipping, E. Europium binding by fulvic acids. Anal. Chim. Acta 1998, 239, 171-180.

(4) Chung, K. H.; Lee, W.; Cho, Y.; Choi, G. S.; Lee, C. W. Comparison of synchronous and laser-induced fluorescence spectroscopy applied to the Eu(III)-fulvate complexation. Talanta 2005,

65, 389-395.

(5) Kerndorff, H.; Schnitzer, M. Sorption of metals on humic acid. Geochim. Cosmochim. Acta

1980, 44, 1701-1708.

(6) Alberts, J. J.; Filip, Z. Metal binding in estuarine humic and fulvic acids: FTIR analysis of humic acid-metal complexes. Environ. Technol. 1998, 19, 923-931.

(7) Pinheiro, J. P.; Mota, A. M.; Benedetti, M. F. Lead and calcium binding to fulvic acids: Salt effect and competition. Environ. Sci. Technol. 1999, 33, 3398-3404.

(8) Hering, J. G.; Morel, F. M. M. Humic acid complexation of calcium and copper. Environ.

12 (9) Hering, J. G.; Morel, F. M. M. Kinetics of trace metal complexation: role of alkaline-earth metals. Environ. Sci. Technol. 1988, 22, 1469-1478.

(10) Tipping, E.; Backes, C. A.; Hurley, M. A. The complexation of protons, aluminum and calcium by aquatic humic substances - A model incorporating binding-site heterogeneity and macroionic effects. Water Res. 1988, 22, 597-611.

(11) Pinheiro, J. P.; Mota, A. M.; Benedetti, M. F. Effect of aluminum competition on lead and cadmium binding to humic acids at variable ionic strength. Environ. Sci. Technol. 2000, 34, 5137-5143.

(12) Kinniburgh, D. G.; van Riemsdijk, W. H.; Koopal, L. K.; Borkovec, M.; Benedetti, M. F.; Avena, M. J. Ion binding to natural organic matter: competition, heterogeneity, stoichiometry and thermodynamic consistency. Colloids Surf., A 1999, 151, 147-166.

(13) Milne, C. J.; Kinniburgh, D. G.; Tipping, E. Generic NICA-Donnan model parameters for proton binding by humic substances. Environ. Sci. Technol. 2001, 35, 2049-2059.

(14) Milne, C. J.; Kinniburgh, D. G.; Van Riemsdijk, W. H.; Tipping, E. Generic NICA-Donnan model parameters for metal-ion binding by humic substances. Environ. Sci. Technol. 2003, 37, 958-971.

(15) Benedetti, M. F. Metal ion binding to colloids from database to field systems. J. Geochem.

Explor. 2006, 88, 81-85.

(16) Weber, T.; Allard, T.; Tipping, E.; Benedetti, M. F. Modeling iron binding to organic matter.

Environ. Sci. Technol. 2006, 40, 7488-7493.

(17) Marang, L.; Reiller, P.; Pepe, M.; Benedetti, M. F. Donnan membrane approach: From equilibrium to dynamic speciation. Environ. Sci. Technol. 2006, 40, 5496-5501.

(18) Marang, L.; Reiller, P.; Jouvin, D.; Benedetti, M. F. U(VI) and Mg(II) binding to humic substances: speciation and competition. Appl. Geochem. submitted.

(19) Wolf, M.; Buckau, G.; Geyer, S. Isolation and characterization of new batches of Gohy-573 humic and fulvic acids. In Humic Substances in Performance Assessment of Nuclear Waste

Disposal: Actinide and Iodine Migration in the Far-Field. Second Technical Progress Report;

Buckau, G., Ed.: Karlsruhe, 2004; pp 111-124. http://bibliothek.fzk.de/zb/berichte/FZKA6969.pdf (20) Buckau, G.; Artinger, R.; Fritz, P.; Geyer, S.; Kim, J. I. Origin and mobility of humic colloids in the Gorleben aquifer system. Appl. Geochem 2000, 15, 171-179.

(21) Baes, C. F.; Mesmer, R. E. The hydrolysis of cations; Wiley Interscience Publication: New-York, 1976.

(22) Hummel, W.; Berner, U.; Curti, E.; Pearson, F. J.; Thoenen, T. Nagra/PSI Chemical Thermodynamic Data Base 01/01, NTB 02-06, NAGRA, Parkland, Florida, 2002.

(23) Keizer, M. G.; Van Riemsdijk, W. H. A Computer Program for the Calculation of Chemical

Speciation and Transport in Soil-Water Systems (ECOSAT 4.7). Agricultural University of

Wageningen: Wageningen, 1994.

(24) Benedetti, M. D. F.; Milne, C. J.; Kinniburgh, D. G.; van Riemsdijk, W. H.; Koopal, L. K. Metal ion binding to humic substances: application of the non-ideal competitive adsorption model.

Environ. Sci. Technol. 1995, 29, 446-457.

(25) Thomason, J. W.; Susetyo, W.; Carreira, L. A. Fluorescence studies of metal humic

complexes with the use of lanthanide ion probe spectroscopy. Appl. Spectrosc. 1996, 50, 401-408. (26) Shin, H. S.; Rhee, S. W.; Lee, B. H.; Moon, C. H. Metal binding sites and partial structures of soil fulvic and humic acids compared: Aided by Eu(III) luminescence spectroscopy and

DEPT/QUAT C-13 NMR pulse techniques. Org. Geochem. 1996, 24, 523-529.

(27) Cabaniss, S. E. Synchronous fluorescence-spectra of metal-fulvic acid complexes. Environ.

Sci. Technol. 1992, 26, 1133-1139.

(28) Avena, M. J.; Koopal, L. K.; van Riemsdijk, W. H. Proton binding to humic acids: Electrostatic and intrinsic interaction. J. Colloid Interface Sci. 1999, 217, 37-48.

(29) Kinniburgh, D. G.; Milne, C. J.; Benedetti, M. D. F.; Pinheiro, J. P.; Filius, J.; van Riemsdijk, W. H. Metal ion binding by humic acid: Application of the NICA-Donnan model.

13 (30) Artinger, R.; Schuessler, W.; Schaefer, T.; Kim, J. I. A kinetic study of Am(III)/humic colloid interactions. Environ. Sci. Technol. 2002, 36, 4358-4363.

(31) Caceci, M.; Moulin, V. Investigations of humic acid samples of different sources by photon correlation spectroscopy. In Lecture Notes in Earth Sciences, Humic Substances in the Aquatic and

Terrestrial Environment; Allard, B., Boren, H., Grimwall, A., Eds.; Lecture Notes in Earth

Sciences; Springer: Berlin, 1991; Vol. 33, pp 97-104.

(32) Yates, L. M.; von Wandruszka, R. Effects of pH and metals on the surface tension of aqueous humic materials. Soil Sci. Soc. Amer. J. 1999, 63, 1645-1649.

14 Table 1. Parameters Derived Using the NICA-Donnan Modela

n1 log K1 ~ n2 log K2 ~ H+ 0.8 4.11 0.65 8.92 Cu2+ 0.6 2.85 0.34 7 Ca2+ 0.78b -1.37c 0.75 c -0.43 c Eu3+ 0.6 1.8 0.36 c 3.43 c a

The intrinsic heterogeneity parameters p for Gohy-573 HA are p1 = 0.8 and p2 = 0.41 and site

density Q1 = 2.63 mol kg-1 and Q2 = 3.08 mol kg-1 (17), b Specific parameters for Gohy-573 (17),

c

15 Figure 1. Cu2+ and Ca2+ binding isotherms to Gohy-573 HA at pH 5.5 with a salt concentration of

I = 1 mM KNO3. The Cu 2+

and Ca2+ binding isotherms to Gohy-573 HA were measured with Cu2+ and Ca2+ ISE, respectively. Cu2+ binding isotherms correspond to a total amount of 60 mg/L Gohy-573 HA with a 10 ≤ [Cu]Total (µM) ≤ 75. Ca2+ binding isotherms correspond to a total amount

of 200 mg /L Gohy-573 HA with 9 ≤ [Ca]Total (µM) ≤ 200. All data points are used to obtain

NICA-Donnan parameters given in Table 1 corresponding to the fitted model line for an ionic strength of 1 or 1.5 mM KNO3. -2 -1 0 1 -7 -6 -5 -4

log[M

]

(mol/L)

lo

g

[M

]

(

m

o

l/k

g

)

16 Figure 2. Eu3+/Cu2+ competitive binding to Gohy-573 HA (a) and Eu3+/Ca2+ competitive binding to Gohy-573 HA (b) at pH 5.5 with a salt concentration of I = 1 mM KNO3. The Cu

2+

and Ca2+ binding to Gohy-573 HA were followed with Cu2+ and Ca2+ ISE, respectively. Eu3+/Cu2+ competitive experiments correspond to a total amount of 60 mg/L Gohy-573 HA with [Cu]Total = 20

µM and 3 ≤ [Eu]Total (µM) ≤ 60. Eu 3+

/Ca2+ competitive experiments correspond to a total amount of 200 mg/L Gohy-573 HA with a [Ca]Total = 30 µM and with 1 ≤ [Eu]Total (µM) ≤ 50. All data points

are used to obtain NICA–Donnan parameters given in Table 1 corresponding to the fitted model line. (a) (µM) 0% 10% 20% 30% 40% 50% 0 20 40 60

[Eu] totalin solution

% of Cu fr e e in solut ion (a) (µM) 0% 10% 20% 30% 40% 50% 0 20 40 60

[Eu] totalin solution

% of Cu fr e e in solut ion (b) (µM) 0% 10% 20% 30% 40% 50% 0 20 40 60

[Eu]totalin solution

% of C a fr ee en s o lut io n (b) (µM) 0% 10% 20% 30% 40% 50% 0 20 40 60

[Eu]totalin solution

% of C a fr ee en s o lut io n (b)

17 Figure 3. Eu3+/Cu2+ competitive binding to Gohy-573 HA (20 mg/L) at pH 5.5 with a salt

concentration of I = 1 mM and with a [Eu]Total = 7 µM and 1 µM ≤ [Cu]Total ≤ 0.3 mM (a).

Eu3+/Ca2+ competitive binding to Gohy-573 HA (15 mg/L) at pH 5.5 with an ionic strength of I = 0.1 M and 0.1 M < I < 0.3 M with [Eu]Total = 5 µM and 1 µM ≤ [Ca]Total ≤ 0.1 M (b).

The Eu binding to Gohy-573 HA was measured with TRLFS. The solid and dashed lines represent the NICA-Donnan prediction calculated with parameters of Table 1. Error bars correspond to a 10% error on TRLFS experiments.

0% 20% 40% 60% 80%

0E+00 1E-04 2E-04 3E-04

[Cu]total in solution (mol/L)

% o f E u f re e in s o lu ti o n 0 0.1 0.2 0.3 mM (a) 0% 20% 40% 60% 80%

0E+00 1E-04 2E-04 3E-04

[Cu]total in solution (mol/L)

% o f E u f re e in s o lu ti o n 0 0.1 0.2 0.3 mM (a) 0% 20% 40% 60% 80%

0E+00 2E-02 4E-02 6E-02 8E-02 1E-01

[Ca]total in solution (mol/L)

% o f E u fr ee i n so lu ti o n Experimental data I = 0.1 M Experimental data 0.1 M < I < 0.3 M I = 0.1 M I = 0.3 M 0 0.02 0.04 0.06 0.08 0.1 (b) 0% 20% 40% 60% 80%

0E+00 2E-02 4E-02 6E-02 8E-02 1E-01

[Ca]total in solution (mol/L)

% o f E u fr ee i n so lu ti o n Experimental data I = 0.1 M Experimental data 0.1 M < I < 0.3 M I = 0.1 M I = 0.3 M 0 0.02 0.04 0.06 0.08 0.1 (b)

18 Figure 4. Measured free Eu3+ concentration versus predicted model values for Eu3+/Cu2+ and Eu3+/Ca2+ competitive binding experiments on Gohy-573 HA. Error bars correspond to a 10% uncertainty on TRLFS experiments.

0E+00 2E-06 4E-06 6E-06

0E+00 2E-06 4E-06 6E-06

[Eu 3+]free measured ( mol.L-1)

[E u 3+ ]free s imu la te d ( mo l. L -1 ) Cu Ca µ µ 0 2 4 6 0 2 4 6 0E+00 2E-06 4E-06 6E-06

0E+00 2E-06 4E-06 6E-06

[Eu 3+]free measured ( mol.L-1)

[E u 3+ ]free s imu la te d ( mo l. L -1 ) Cu Ca µ µ 0 2 4 6 0 2 4 6

19 Figure 5. Eu3+, Ca2+, and Cu2+ individual calculated speciation. The calculations are made with a Gohy-573 HA concentration of 20 mg/L and a salt concentration of 1mM for [Eu]Total = 7 µM (a)

and for [Cu]Total = 1 µM (b). The calculations are made with a Gohy-573 HA concentration of 15

mg/L and a salt concentration of 0.1 M for [Eu]Total = 5 µM (c) and for [Cu]Total = 1 mM (d). The

specific binding sites ( ) ( ) correspond to the carboxylic type of sites and phenolic type of sites, respectively. ( ) corresponds to the metal ion located in the humic Donnan phase.

Eu speciation I= 0.001 M S1 S2 Donnan Eu speciation I=0.1 M Cu speciation I= 0.001 M Ca speciation I= 0.1 M (a) (b) (c) _(d)